Detailed observations from TF-SIF, the Coast Guard Dash 9 were performed in the afternoon of the 23rd of August. The observations revealed that no signs of an ongoing eruption could be detected and it is concluded that a subglacial eruption did not happen. Increased low frequency tremor observed on that day must therefore have a different explanation.
The SAR-radar images obtained by the aircraft in the afternoon of 23 August show that no changes have occurred, neither subsidence, or unusal discharge of rivers.
Experience from subglacial eruptions in the past suggests the following:
· Large scale melting of ice occurs if an eruption takes place beneath a glacier. Even very small eruptions (like the eruption at Fimmvorðuháls that preceded the Eyjafjallajökull eruption in 2010) if they had taken place under ice, would melt of the order of 100 cubic meters per second of ice and cause notable increase in discharge in glacial rivers.
· In most cases meltwater will flow away from the eruption site towards the edge of the glacier. If an eruption occurs in late summer the drainage system at the bottom of the glacier is well developed. This means that water will drain fast to the edge of the glacier. For example, if an eruption site is located under ice, 5-10 km from the edge of the glacier it is expected that meltwater will reach the edge in about an hour.
· The time it takes an eruption to melt its way to the surface is strongly dependent on the ice thickness. As an example, the moderate sized eruption of Grímsvötn in 2004, melted through 150 m of ice in about one hour while the more powerful Gjálp eruption in 1996 took 31 hours to melt through 600 m of ice.
· The above implies that the meltwater from an eruption beginning under e.g. 500 m of ice in Dyngjujökull would emerge from the glacier several hours before the eruption melts through the glacier sparking off an explosive eruption because of magma-water interaction.
Hazard due to eruptions, including subglacial ones (Jökull 2008):
On the volcano-ice interaction in the Gjálp eruption:
And on eruptions in glaciers in general: